Tune crashes with DataModule

[lukasfolle]

🐛 Bug

Connected to #1495.
When using trainer.tune() only the model is forwarded, but no the datamodule cp.: https://github.com/PyTorchLightning/pytorch-lightning/blob/eafec7d425bf691ab5bfaf2794c3e581487ecfa8/pytorch_lightning/tuner/tuning.py#L64

Expected behavior

Just like for self.trainer.auto_scale_batch_size the datamodule or train/val dataloaders should be forwarded as well.
I.e.:

self.lr_find(
  model,
  train_dataloader=train_dataloader,
  val_dataloaders=val_dataloaders,
  datamodule=datamodule, ...)

[SkafteNicki]

Fixed by #6784